As of the end of January 1997 Phoenix has been used in four shared-risk user runs. The weather in January has not been kind with nearly 75% of the time lost to bad weather. These runs plus two test and evaluation runs in fall 1996 have produced much additional information on the instrument since the last Newsletter. Due to the complexity of the instrument, there is still a lot that is not known. Overall, the shared-risk users have been excited by the performance of the instrument. In the figure we present a high quality spectrum that resulted from the shared-risk time.
We are continuing to debug the instrument. In particular, while the dispersion is 1.5 km/s per pixel, the current resolution is 5 km/s (R = 60000, full width at half maximum of unresolved absorption lines), somewhat less than the desired resolution of 3 km/s over 2 pixels (R = 100000). During the next year we expect to work on understanding the collimator focus and imaging.
There has been insufficient test time to adequately determine limiting magnitudes throughout the range of operating wavelengths. The integration time in the 35µm region is limited by the sky spectrum that appears in emission. Because this background is thermal, its level, and thus the maximum integration time, is a function of the ambient temperature. During the September 1996 engineering run, the maximum integration time was 1 minute at 4.6µm. Stars as faint as 4th magnitude have been successfully observed at 4.6µm. At 3.6µm the maximum integration time in January was 20 minutes. In the 1-2.5µm region we observe a background due to a light leak of roughly 1.5 ADU/s. We intend to attack this annoying problem aggressively. Limiting magnitude for high signal-to-noise in the K band appears to be in the 8-9th magnitude range, but still has not been tested in detail. At the short wavelength range of performance, the limiting magnitude at 1µm is in the 6-7 magnitude range. The decreased performance in this region is due in part to a falloff in reflectance of the visual/infrared dichroic.
A manual is being prepared. We plan to update the Phoenix web site http://www.noao.edu/kpno/phoenix/phoenix.html with this manual and a current set of tables and illustrations. A test run in late January at the 4-m should provide additional information on the performance.
Phoenix will be available on both the 2.1-m and 4-m telescopes in the fall 1997 semester. Users wishing to use the 4-m should justify this choice. Two pixels on Phoenix are 0.7" on the 2.1-m but only 0.4" on the 4m. The widest slit available in Phoenix is 4 pixels. Given typical seeing of ~ 0.8" in the 2µm region, a gain of roughly 1 magnitude in limiting performance in the 1-2µm region is expected with the widest slit. However, this gain will not be realized in poor seeing or with the more narrow slits. Also, since the scale is smaller at the 4-m, the slit is proportionately shorter, about 30", compared to 1' at the 2.1-m. The other significant difference between these telescopes is that the 4-m permits instrument rotation and offset guiding over a wide field. Therefore, projects with optically invisible targets, but visible stars within the 4-m guider field, would benefit from use on this telescope.
Phoenix has an internal dichroic that permits viewing of the on-axis field in visible light using a CCD camera mounted on the instrument. A mirror can be folded into the beam behind the slit to image the field in the infrared. The slit can be removed to view the entire field, about 1' diameter. The acquisition procedure is to acquire the infrared source through the open aperture, rotate the observing slit into place and center the transmitted image, and then guide on a visible object in the CCD camera field. Tests show that flexure between the visual and infrared image is not measurable at the one pixel level.
The NOAO long range plan calls for Phoenix to be sent to CTIO for two years starting in July 1998. Future shared use with Gemini is also expected.
Additional information can be found on the web site (http://www.noao.edu/kpno/phoenix/phoenix.html) or by contacting email@example.com.
Ken Hinkle, Dick Joyce